diff --git a/python/cudf/cudf/core/_base_index.py b/python/cudf/cudf/core/_base_index.py
index d56345a133d..3616ec1b542 100644
--- a/python/cudf/cudf/core/_base_index.py
+++ b/python/cudf/cudf/core/_base_index.py
@@ -1836,7 +1836,7 @@ def __array_function__(self, func, types, args, kwargs):
             return NotImplemented
 
     @classmethod
-    def from_pandas(cls, index, nan_as_null=None):
+    def from_pandas(cls, index, nan_as_null=no_default):
         """
         Convert from a Pandas Index.
 
@@ -1866,6 +1866,11 @@ def from_pandas(cls, index, nan_as_null=None):
         >>> cudf.Index.from_pandas(pdi, nan_as_null=False)
         Float64Index([10.0, 20.0, 30.0, nan], dtype='float64')
         """
+        if nan_as_null is no_default:
+            nan_as_null = (
+                False if cudf.get_option("mode.pandas_compatible") else None
+            )
+
         if not isinstance(index, pd.Index):
             raise TypeError("not a pandas.Index")
 
diff --git a/python/cudf/cudf/core/dataframe.py b/python/cudf/cudf/core/dataframe.py
index 89e1e58516a..46c7557148e 100644
--- a/python/cudf/cudf/core/dataframe.py
+++ b/python/cudf/cudf/core/dataframe.py
@@ -5184,7 +5184,7 @@ def to_pandas(self, nullable=False, **kwargs):
 
     @classmethod
     @_cudf_nvtx_annotate
-    def from_pandas(cls, dataframe, nan_as_null=None):
+    def from_pandas(cls, dataframe, nan_as_null=no_default):
         """
         Convert from a Pandas DataFrame.
 
@@ -5213,6 +5213,11 @@ def from_pandas(cls, dataframe, nan_as_null=None):
         1  1  2
         2  3  4
         """
+        if nan_as_null is no_default:
+            nan_as_null = (
+                False if cudf.get_option("mode.pandas_compatible") else None
+            )
+
         if isinstance(dataframe, pd.DataFrame):
 
             if not dataframe.columns.is_unique:
@@ -7770,7 +7775,7 @@ def func(left, right, output):
 
 
 @_cudf_nvtx_annotate
-def from_pandas(obj, nan_as_null=None):
+def from_pandas(obj, nan_as_null=no_default):
     """
     Convert certain Pandas objects into the cudf equivalent.
 
@@ -7868,6 +7873,11 @@ def from_pandas(obj, nan_as_null=None):
     >>> type(pmidx)
     <class 'pandas.core.indexes.multi.MultiIndex'>
     """
+    if nan_as_null is no_default:
+        nan_as_null = (
+            False if cudf.get_option("mode.pandas_compatible") else None
+        )
+
     if isinstance(obj, pd.DataFrame):
         return DataFrame.from_pandas(obj, nan_as_null=nan_as_null)
     elif isinstance(obj, pd.Series):
diff --git a/python/cudf/cudf/core/multiindex.py b/python/cudf/cudf/core/multiindex.py
index 8c05c7a328e..c904d1b021b 100644
--- a/python/cudf/cudf/core/multiindex.py
+++ b/python/cudf/cudf/core/multiindex.py
@@ -1604,7 +1604,7 @@ def to_pandas(self, nullable=False, **kwargs):
 
     @classmethod
     @_cudf_nvtx_annotate
-    def from_pandas(cls, multiindex, nan_as_null=None):
+    def from_pandas(cls, multiindex, nan_as_null=no_default):
         """
         Convert from a Pandas MultiIndex
 
@@ -1625,6 +1625,10 @@ def from_pandas(cls, multiindex, nan_as_null=None):
         """
         if not isinstance(multiindex, pd.MultiIndex):
             raise TypeError("not a pandas.MultiIndex")
+        if nan_as_null is no_default:
+            nan_as_null = (
+                False if cudf.get_option("mode.pandas_compatible") else None
+            )
 
         # if `multiindex` has two or more levels that
         # have the same name, then `multiindex.to_frame()`
diff --git a/python/cudf/cudf/core/series.py b/python/cudf/cudf/core/series.py
index cadd53fa516..66e8d00a934 100644
--- a/python/cudf/cudf/core/series.py
+++ b/python/cudf/cudf/core/series.py
@@ -688,7 +688,7 @@ def __contains__(self, item):
 
     @classmethod
     @_cudf_nvtx_annotate
-    def from_pandas(cls, s, nan_as_null=None):
+    def from_pandas(cls, s, nan_as_null=no_default):
         """
         Convert from a Pandas Series.
 
@@ -726,6 +726,10 @@ def from_pandas(cls, s, nan_as_null=None):
         3     NaN
         dtype: float64
         """
+        if nan_as_null is no_default:
+            nan_as_null = (
+                False if cudf.get_option("mode.pandas_compatible") else None
+            )
         with warnings.catch_warnings():
             warnings.simplefilter("ignore")
             result = cls(s, nan_as_null=nan_as_null)
diff --git a/python/cudf/cudf/tests/test_series.py b/python/cudf/cudf/tests/test_series.py
index b0cd9937835..e606c36e855 100644
--- a/python/cudf/cudf/tests/test_series.py
+++ b/python/cudf/cudf/tests/test_series.py
@@ -2575,3 +2575,10 @@ def test_series_categorical_missing_value_count():
     actual = gs.value_counts()
 
     assert_eq(expected, actual, check_dtype=False)
+
+
+def test_series_error_nan_mixed_types():
+    ps = pd.Series([np.nan, "ab", "cd"])
+    with cudf.option_context("mode.pandas_compatible", True):
+        with pytest.raises(pa.ArrowInvalid):
+            cudf.from_pandas(ps)